Large animals are necessary models to test TMJ protheses, to test the use of cartilage and disc grafts, and in studies on experimentally induced osteoarthritis. Pigs and sheep are now the animals of choice, both for practical reasons (cost and ease of use) and because their TMJ components are similar in size to those of humans. Indeed, a consideration of muscle and occlusal anatomy suggests that TMJ mechanics of pigs and sheep have very little in common. Many of the TMJ studies on pigs and sheep are affected by joint mechanics, yet there is little information or appreciation of the differences and similarities to the human condition. In the proposed study we will explore these differences and similarities by measuring three-dimensional loads on the mandible during masticatory function in pigs and sheep. Specifically we will assess torsion and bending of the condyle and mandibular corpus (S.A. 1) by measuring patterns of strain from their medial and lateral surfaces. We expect a three-dimensional analysis of strain will reveal complex patterns of mandibular deformation in both species. Because pigs have balancing side occlusal contacts and a fused symphysis and sheep have working-side only occlusal contact and an open symphysis, we expect dramatic species differences in the relative magnitudes of balancing versus working side condylar and corporal strain. We will also measure the magnitudes and orientations of occlusal forces using a miniature three-component force transducer implanted under the crown of a molar tooth (S.A. 2). Because of the technically challenging nature of this part of the proposed research, we will begin by using it on high-crowned sheep molars. Although sheep show a large medial translation of their mandible during the chew cycle, we expect that their flat-occlusal surface will provide little resistance in this direction. We will simultaneously record condylar and corporal strain and EMG activity, allowing us to explore how strain patterns and occlusal forces change in relation to each other and in relation to muscle activation during the course of the power stroke. The data we generate will be used in comparison with finite element predictions of mandibular deformations.